February 2024 - U-Boot

[PATCH] bootflow: Fix build error when BOOTMETH_CROS is disabled
by Dan Carpenter 04 Mar '24

04 Mar '24

The bootflow testing assumes that BOOTMETH_CROS is enabled but it might not be which leads to a build error. aarch64-linux-gnu-ld.bfd: test/boot/bootflow.o: in function `prep_mmc_bootdev': test/boot/bootflow.c:549:(.text.prep_mmc_bootdev+0x1c8): undefined reference to `_u_boot_list_2_driver_2_bootmeth_cros' Fixes: d08db02d2d3d ("bootstd: Add a test for bootmeth_cros") Signed-off-by: Dan Carpenter <dan.carpenter(a)linaro.org> --- test/boot/bootflow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test/boot/bootflow.c b/test/boot/bootflow.c index fa54dde661c8..4845b7121c84 100644 --- a/test/boot/bootflow.c +++ b/test/boot/bootflow.c @@ -544,7 +544,7 @@ static int prep_mmc_bootdev(struct unit_test_state *uts, const char *mmc_dev, "bootmeth_script", 0, ofnode_null(), &dev)); /* Enable the cros bootmeth if needed */ - if (bind_cros) { + if (IS_ENABLED(CONFIG_BOOTMETH_CROS) && bind_cros) { ut_assertok(uclass_first_device_err(UCLASS_BOOTSTD, &bootstd)); ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_cros), "cros", 0, ofnode_null(), &dev)); -- 2.43.0

2 1

[PATCH] boot: Only define checksum algos when the hashes are enabled
by Sean Anderson 04 Mar '24

04 Mar '24

Don't define checksum algos when the underlying hashes are not enabled. This allows disabling these hashes in SPL (or U-Boot). Fixes: d16b38f4270 ("Add support for SHA384 and SHA512") Fixes: 646257d1f40 ("rsa: add sha256-rsa2048 algorithm") Signed-off-by: Sean Anderson <sean.anderson(a)seco.com> --- boot/image-sig.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/boot/image-sig.c b/boot/image-sig.c index b5692d58b24..0421a61b040 100644 --- a/boot/image-sig.c +++ b/boot/image-sig.c @@ -17,6 +17,7 @@ DECLARE_GLOBAL_DATA_PTR; #define IMAGE_MAX_HASHED_NODES 100 struct checksum_algo checksum_algos[] = { +#if CONFIG_IS_ENABLED(SHA1) { .name = "sha1", .checksum_len = SHA1_SUM_LEN, @@ -24,6 +25,8 @@ struct checksum_algo checksum_algos[] = { .der_prefix = sha1_der_prefix, .calculate = hash_calculate, }, +#endif +#if CONFIG_IS_ENABLED(SHA256) { .name = "sha256", .checksum_len = SHA256_SUM_LEN, @@ -31,7 +34,8 @@ struct checksum_algo checksum_algos[] = { .der_prefix = sha256_der_prefix, .calculate = hash_calculate, }, -#ifdef CONFIG_SHA384 +#endif +#if CONFIG_IS_ENABLED(SHA384) { .name = "sha384", .checksum_len = SHA384_SUM_LEN, @@ -40,7 +44,7 @@ struct checksum_algo checksum_algos[] = { .calculate = hash_calculate, }, #endif -#ifdef CONFIG_SHA512 +#if CONFIG_IS_ENABLED(SHA512) { .name = "sha512", .checksum_len = SHA512_SUM_LEN, -- 2.35.1.1320.gc452695387.dirty [Embedded World 2024, SECO SpA]<https://www.messe-ticket.de/Nuernberg/embeddedworld2024/Register/ew24517689>

2 1

[PATCH] bootdev: drop unnecessary assert on bootflow->bdev
by Thomas Weißschuh 04 Mar '24

04 Mar '24

Not all flows have a device and the function already contains logic to handle this case. Fixes: eccb25cd5922 ("bootstd: Allow the bootdev to be optional in bootflows") Signed-off-by: Thomas Weißschuh <thomas.weissschuh(a)linutronix.de> --- boot/bootdev-uclass.c | 1 - 1 file changed, 1 deletion(-) diff --git a/boot/bootdev-uclass.c b/boot/bootdev-uclass.c index 35afb93c0e73..0fa6dad8b11b 100644 --- a/boot/bootdev-uclass.c +++ b/boot/bootdev-uclass.c @@ -39,7 +39,6 @@ int bootdev_add_bootflow(struct bootflow *bflow) struct bootflow *new; int ret; - assert(bflow->dev); ret = bootstd_get_priv(&std); if (ret) return ret; --- base-commit: e8f2404e093daf6cc3ac2b3233e3c6770d13e371 change-id: 20240209-bootdev-assert-11714f63d72d Best regards, -- Thomas Weißschuh <thomas.weissschuh(a)linutronix.de>

2 1

[PATCH v2 0/1] Fix booting kernels with ATAGS and extlinux
by Svyatoslav Ryhel 04 Mar '24

04 Mar '24

Currently, if boot with extlinux.conf and do not set the fdt U-Boot will provide its own device tree. This behavior is beneficial if the U-Boot device tree is in sync with Linux, but it totally halts the booting of pre-dtb kernels (3.4 for example) since it uses ATAGs. To fix this, pass `-` in the fdt extlinux field as a signal that no tree should be used. Tested with 3.4 legacy kernel and mainline 6.7 kernel on P895 (lg_x3 board). --- Changes form v1 - Added CONFIG guards - Clarified documentation --- Svyatoslav Ryhel (1): boot: pxe_utils: skip fdt setup in case legacy kernel is booted boot/pxe_utils.c | 27 ++++++++++++++++++++++----- doc/develop/distro.rst | 6 ++++++ 2 files changed, 28 insertions(+), 5 deletions(-) -- 2.40.1

2 4

[PATCH] test: cmd: mbr: Fix Smatch static checker warning
by Alexander Gendin 04 Mar '24

04 Mar '24

This patch fixes Smatch static checker warning: test/cmd/mbr.c:243 mbr_test_run() warn: sizeof(NUMBER)? Reported-by: Dan Carpenter <dan.carpenter(a)linaro.org> Signed-off-by: Alexander Gendin <agendin(a)matrox.com> --- test/cmd/mbr.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/test/cmd/mbr.c b/test/cmd/mbr.c index 46b78e706c..235b363290 100644 --- a/test/cmd/mbr.c +++ b/test/cmd/mbr.c @@ -240,7 +240,11 @@ static int mbr_test_run(struct unit_test_state *uts) ut_assert(ofnode_valid(node)); ut_assertok(lists_bind_fdt(gd->dm_root, node, &dev, NULL, false)); - mbr_parts_max = sizeof('\0') + 2 + + /* + * 1 byte for null character + * 2 reserved bytes + */ + mbr_parts_max = 1 + 2 + strlen(mbr_parts_header) + strlen(mbr_parts_p1) + strlen(mbr_parts_p2) + -- 2.41.0

3 2

[PATCH v5] test/py: net: Add dhcp abort test
by Love Kumar 04 Mar '24

04 Mar '24

Abort the dhcp request in the middle by pressing ctrl + c on u-boot prompt and validate the abort status. Signed-off-by: Love Kumar <love.kumar(a)amd.com> --- Changes in v2: - Mark CMD_MII command dependency Changes in v3: - Skip the test if PHY device not present Changes in v4: - Setup the network configuration by running dhcp test at the end of test - Add option to skip the test if it is not desired or possible Changes in v5: - Disable this test case by default - Use single quote to maintain consistency - Setup the network configuration before exiting the test --- test/py/tests/test_net.py | 58 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 58 insertions(+) diff --git a/test/py/tests/test_net.py b/test/py/tests/test_net.py index 4ff3dafd6293..038a473b2394 100644 --- a/test/py/tests/test_net.py +++ b/test/py/tests/test_net.py @@ -8,6 +8,7 @@ import pytest import u_boot_utils import uuid import datetime +import re """ Note: This test relies on boardenv_* containing configuration values to define @@ -31,6 +32,12 @@ env__net_uses_pci = True # set to False. env__net_dhcp_server = True +# False or omitted if a DHCP server is attached to the network, and dhcp abort +# case should be tested. +# If DHCP abort testing is not possible or desired, set this variable to True. +# For example: On some setup, dhcp is too fast and this case may not work. +env__dhcp_abort_test_skip = True + # True if a DHCPv6 server is attached to the network, and should be tested. # If DHCPv6 testing is not possible or desired, this variable may be omitted or # set to False. @@ -120,6 +127,57 @@ def test_net_dhcp(u_boot_console): global net_set_up net_set_up = True +(a)pytest.mark.buildconfigspec('cmd_dhcp') +(a)pytest.mark.buildconfigspec('cmd_mii') +def test_net_dhcp_abort(u_boot_console): + """Test the dhcp command by pressing ctrl+c in the middle of dhcp request + + The boardenv_* file may be used to enable/disable this test; see the + comment at the beginning of this file. + """ + + test_dhcp = u_boot_console.config.env.get('env__net_dhcp_server', False) + if not test_dhcp: + pytest.skip('No DHCP server available') + + if u_boot_console.config.env.get('env__dhcp_abort_test_skip', True): + pytest.skip('DHCP abort test is not enabled!') + + u_boot_console.run_command('setenv autoload no') + + # Phy reset before running dhcp command + output = u_boot_console.run_command('mii device') + if not re.search(r"Current device: '(.+?)'", output): + pytest.skip('PHY device does not exist!') + eth_num = re.search(r"Current device: '(.+?)'", output).groups()[0] + u_boot_console.run_command(f'mii device {eth_num}') + output = u_boot_console.run_command('mii info') + eth_addr = hex(int(re.search(r'PHY (.+?):', output).groups()[0], 16)) + u_boot_console.run_command(f'mii modify {eth_addr} 0 0x8000 0x8000') + + u_boot_console.run_command('dhcp', wait_for_prompt=False) + try: + u_boot_console.wait_for('Waiting for PHY auto negotiation to complete') + except: + pytest.skip('Timeout waiting for PHY auto negotiation to complete') + + u_boot_console.wait_for('done') + + try: + # Sending Ctrl-C + output = u_boot_console.run_command( + chr(3), wait_for_echo=False, send_nl=False + ) + assert 'TIMEOUT' not in output + assert 'DHCP client bound to address ' not in output + assert 'Abort' in output + finally: + # Provide a time to recover from Abort - if it is not performed + # There is message like: ethernet@ff0e0000: No link. + u_boot_console.run_command('sleep 1') + # Run the dhcp test to setup the network configuration + test_net_dhcp(u_boot_console) + @pytest.mark.buildconfigspec('cmd_dhcp6') def test_net_dhcp6(u_boot_console): """Test the dhcp6 command. -- 2.25.1

2 1

[PATCH v3 0/2] arm: s5p4418: fix relocation
by Stefan Bosch 04 Mar '24

04 Mar '24

Fix relocation of u-boot for s5p4418-SoC. I.e. use __image_copy_start instead of _start to calculate the monitor length. Furthermore use an adapted version of relocate_vectors for the s5p4418-SoC. Background: The header (NSIH) used by the 2nd-bootloader (included at the begin of u-boot.bin) is not loaded into RAM. Therefore _start has to be after the header and therefore is not equal to __image_copy_start which is at the begin of the header. Changes in v3: - MICROBLAZE has not built anymore. Therefore do change calculation of gd->mon_len for ARCH_NEXELL only. This makes sure that MICROBLAZE and other ARM-boards are not affected. Changes in v2: - Cosmetic: Fix spelling mistake in commit message Stefan Bosch (2): common: board_f: change calculation of gd->mon_len to fix s5p4418 reloc arm: s5p4418: fix relocation of vectors arch/arm/cpu/armv7/s5p4418/Makefile | 3 +++ arch/arm/cpu/armv7/s5p4418/relocate.S | 24 ++++++++++++++++++++++++ common/board_f.c | 4 +++- 3 files changed, 30 insertions(+), 1 deletion(-) create mode 100644 arch/arm/cpu/armv7/s5p4418/relocate.S -- 2.17.1

3 7

[PATCH v2] qemu-arm: round down memory to multiple of 2MB for LPAE
by Igor Opaniuk 04 Mar '24

04 Mar '24

QEMU's -m option can take fractional megabyte values, and lowest granularity seems to be 0x2000. For example, run qemu with amount of memory set to 100005k (0x61A9400): $ qemu-system-arm -machine virt -cpu cortex-a15 -m 100005k \ -bios denx/u-boot.bin -nographic => fdt addr $fdt_addr => fdt print /memory@40000000 memory@40000000 { reg = <0x00000000 0x40000000 0x00000000 0x061aa000>; device_type = "memory"; }; When LPAE is enabled, 1:1 mapping is created using 2 MB blocks. In case amount of memory provided to QEMU is not multiple of 2 MB, hang occurs during MMU initialization. How to reproduce: qemu-system-arm -machine virt -m 1058 -nographic -bios u-boot.bin - boots qemu-system-arm -machine virt -m 1057 -nographic -bios u-boot.bin - hangs DRAM: 1 GiB initcall: 60011df8 initcall: 60011904 New Stack Pointer is: 80fffe90 initcall: 60011a20 initcall: 60011bcc initcall: 60011bd4 initcall: 600119b4 Relocation Offset is: 22042000 Relocating to 82042000, new gd at 81001ed0, sp at 80fffe90 initcall: 60011b8c initcall: 82053ea0 initcall: 82053ea8 initcall: 60012040 (relocated to 82054040) dram_bank_mmu_setup: bank: 0 - hang here during mmu init - This patches rounds down to the nearest multiple of 2MB when CONFIG_ARMV7_LPAE=y. Fixes: 3fa914af82("arm: qemu: implement enable_caches()") Signed-off-by: Igor Opaniuk <igor.opaniuk(a)foundries.io> --- Changes in v2: - Adjust commit message, add more details board/emulation/qemu-arm/qemu-arm.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/board/emulation/qemu-arm/qemu-arm.c b/board/emulation/qemu-arm/qemu-arm.c index 942f1fff571..ecfd19f1a7e 100644 --- a/board/emulation/qemu-arm/qemu-arm.c +++ b/board/emulation/qemu-arm/qemu-arm.c @@ -127,6 +127,18 @@ int dram_init(void) if (fdtdec_setup_mem_size_base() != 0) return -EINVAL; + /* + * When LPAE is enabled (ARMv7), + * 1:1 mapping is created using 2 MB blocks. + * + * In case amount of memory provided to QEMU + * is not multiple of 2 MB, round down the amount + * of available memory to avoid hang during MMU + * initialization. + */ + if (CONFIG_IS_ENABLED(ARMV7_LPAE)) + gd->ram_size -= (gd->ram_size % 0x200000); + return 0; } -- 2.34.1

3 2

[PATCH] xen: xenguest_arm64: map all VIRTIO MMIO region
by Peng Fan (OSS) 04 Mar '24

04 Mar '24

From: Peng Fan <peng.fan(a)nxp.com> When run `virtio scan` on i.MX95, there is abort when accessing virtio mmio region. The issue is the mmio region is not mapped. So let's map all virtio mmio regions. Signed-off-by: Peng Fan <peng.fan(a)nxp.com> --- board/xen/xenguest_arm64/xenguest_arm64.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/board/xen/xenguest_arm64/xenguest_arm64.c b/board/xen/xenguest_arm64/xenguest_arm64.c index 244070a242d..1d2946f4fde 100644 --- a/board/xen/xenguest_arm64/xenguest_arm64.c +++ b/board/xen/xenguest_arm64/xenguest_arm64.c @@ -31,6 +31,9 @@ DECLARE_GLOBAL_DATA_PTR; +#define GUEST_VIRTIO_MMIO_BASE 0x2000000 +#define GUEST_VIRTIO_MMIO_SIZE 0x100000 + int board_init(void) { return 0; @@ -212,6 +215,15 @@ static int setup_mem_map(void) PTE_BLOCK_INNER_SHARE); i++; + if (CONFIG_IS_ENABLED(VIRTIO_MMIO)) { + xen_mem_map[i].virt = GUEST_VIRTIO_MMIO_BASE; + xen_mem_map[i].phys = GUEST_VIRTIO_MMIO_BASE; + xen_mem_map[i].size = GUEST_VIRTIO_MMIO_SIZE; + xen_mem_map[i].attrs = (PTE_BLOCK_MEMTYPE(MT_DEVICE_NGNRNE) | + PTE_BLOCK_NON_SHARE); + i++; + } + mem = get_next_memory_node(blob, -1); if (mem < 0) { printf("%s: Missing /memory node\n", __func__); @@ -219,6 +231,11 @@ static int setup_mem_map(void) } for (; i < MAX_MEM_MAP_REGIONS; i++) { + if (CONFIG_IS_ENABLED(VIRTIO_MMIO)) { + ret = fdt_node_check_compatible(blob, mem, "virtio,mmio"); + if (!ret) + continue; + } ret = fdt_get_resource(blob, mem, "reg", reg++, &res); if (ret == -FDT_ERR_NOTFOUND) { reg = 0; -- 2.35.3

2 1

[PATCH 03/60] net: mediatek: add support for XGMII interface
by Weijie Gao 04 Mar '24

04 Mar '24

This patch add XGMII support for connecting 2.5G PHY. Signed-off-by: Bo-Cun Chen <bc-bocun.chen(a)mediatek.com> Signed-off-by: Weijie Gao <weijie.gao(a)mediatek.com> --- drivers/net/mtk_eth.c | 22 ++++++++++++++-------- drivers/net/mtk_eth.h | 2 +- 2 files changed, 15 insertions(+), 9 deletions(-) diff --git a/drivers/net/mtk_eth.c b/drivers/net/mtk_eth.c index 726aedad3f..75e7bcf83b 100644 --- a/drivers/net/mtk_eth.c +++ b/drivers/net/mtk_eth.c @@ -1246,7 +1246,8 @@ static int mtk_phy_start(struct mtk_eth_priv *priv) } if (!priv->force_mode) { - if (priv->phy_interface == PHY_INTERFACE_MODE_USXGMII) + if (priv->phy_interface == PHY_INTERFACE_MODE_USXGMII || + priv->phy_interface == PHY_INTERFACE_MODE_XGMII) mtk_xphy_link_adjust(priv); else mtk_phy_link_adjust(priv); @@ -1516,7 +1517,7 @@ static void mtk_mac_init(struct mtk_eth_priv *priv) static void mtk_xmac_init(struct mtk_eth_priv *priv) { - u32 sts; + u32 force_link = 0; switch (priv->phy_interface) { case PHY_INTERFACE_MODE_USXGMII: @@ -1531,15 +1532,19 @@ static void mtk_xmac_init(struct mtk_eth_priv *priv) SYSCFG0_GE_MODE_M << SYSCFG0_GE_MODE_S(priv->gmac_id), 0); - if (priv->gmac_id == 1) { + if (priv->phy_interface == PHY_INTERFACE_MODE_USXGMII && + priv->gmac_id == 1) { mtk_infra_rmw(priv, TOPMISC_NETSYS_PCS_MUX, NETSYS_PCS_MUX_MASK, MUX_G2_USXGMII_SEL); - } else if (priv->gmac_id == 2) { - sts = mtk_gmac_read(priv, XGMAC_STS(priv->gmac_id)); - sts |= XGMAC_FORCE_LINK; - mtk_gmac_write(priv, XGMAC_STS(priv->gmac_id), sts); } + if (priv->phy_interface == PHY_INTERFACE_MODE_XGMII || + priv->gmac_id == 2) + force_link = XGMAC_FORCE_LINK(priv->gmac_id); + + mtk_gmac_rmw(priv, XGMAC_STS(priv->gmac_id), + XGMAC_FORCE_LINK(priv->gmac_id), force_link); + /* Force GMAC link down */ mtk_gmac_write(priv, GMAC_PORT_MCR(priv->gmac_id), FORCE_MODE); } @@ -1828,7 +1833,8 @@ static int mtk_eth_probe(struct udevice *dev) mtk_eth_mdc_init(priv); /* Set MAC mode */ - if (priv->phy_interface == PHY_INTERFACE_MODE_USXGMII) + if (priv->phy_interface == PHY_INTERFACE_MODE_USXGMII || + priv->phy_interface == PHY_INTERFACE_MODE_XGMII) mtk_xmac_init(priv); else mtk_mac_init(priv); diff --git a/drivers/net/mtk_eth.h b/drivers/net/mtk_eth.h index 45229c0f9a..fd31c782c7 100644 --- a/drivers/net/mtk_eth.h +++ b/drivers/net/mtk_eth.h @@ -268,7 +268,7 @@ enum mkt_eth_capabilities { /* XGMAC Status Registers */ #define XGMAC_STS(x) (((x) == 2) ? 0x001C : 0x000C) -#define XGMAC_FORCE_LINK BIT(15) +#define XGMAC_FORCE_LINK(x) (((x) == 1) ? BIT(31) : BIT(15)) /* XGMAC Registers */ #define XGMAC_PORT_MCR(x) (0x2000 + (((x) - 1) * 0x1000)) -- 2.34.1

2 1